FinCast enables financial forecasting of stock prices and cryptocurrencies. For this project, we have used stock market data and cryptocurrency historical data from Yahoo Finance.
We suggest to set up a virtual environment in the hacklytics folder by doing the following:
virtualenv --python python3 venv
To activate it, simply run
source venv/bin/activate
To run the application, you will need to run
pip install -r requirments.txt
When you have installed all the required libraries, simply run
streamlink run web_interface.py
To load the data, we have the function:
import yfinance as yf
# getting latest data
def get_data(stocks, start, end):
df = pdr.get_data_yahoo(stocks, start, end)
return df
The machine learning model to predict stock prices can be found in the notebook below:
./Notebook/FinCast.ipynb
The machine learning model to predict cryptocurrency prices can be found in the notebook below:
./Notebook/CryptoFinCast.ipynb
We have used long short-term memory networks to work on this project.